TL;DR:

  • Only 25% of AI initiatives deliver expected ROI over past three years, with just 18% achieving ROI above cost of capital
  • One in three businesses paused AI deployment after pilot phase due to inadequate infrastructure
  • Hybrid-by-design architecture and strategic partnerships differentiate top-performing organisations

Despite AI adoption reaching unprecedented levels with 78% of companies using AI in at least one business function and a global market valued over $244 billion USD, most organisations struggle to demonstrate financial benefits to stakeholders.

The ROI Reality Gap

Over the past three years, CEOs report that only 25% of their AI initiatives delivered the expected ROI. Research reveals approximately 18% of companies have achieved an ROI above the cost of capital, leaving a significant gap between investment and returns.

The question facing most organisations is no longer “do you use AI?” but rather “how do you use it?”—and whether that usage translates into measurable business value.

Infrastructure as the Bottleneck

Recent IBM Institute for Business Value research found that one in three businesses paused an AI deployment after the initial pilot phase. CTOs and CIOs are realising that AI doesn’t exist in a vacuum but resides in data centres, in the cloud, and at the edge with exponentially growing power demands.

Without proper infrastructure, organisations cannot scale deployments beyond one-off use cases with limited ROI. The difference between high-performing and struggling organisations comes down to infrastructure approach: hybrid-by-design versus hybrid-by-default.

Hybrid-by-Design Architecture

Organisations achieving ROI take a platform approach to AI that provides common capabilities across multiple IT environments. This “hybrid-by-design” architecture optimizes infrastructure for consistency, scale, trust, and performance rather than allowing independently designed environments to emerge opportunistically.

Hybrid-by-default architectures are riddled with limitations, slower innovation, and weaker data management. For example, a financial services company using AI chatbots for customer service may see limited gains if the chatbot draws from a narrow data pool. With enterprise-wide AI infrastructure enabling data-sharing across departments, the same company could unlock dramatically more value.

Security and Compliance Imperatives

IBM’s latest Cost of a Data Breach report reveals the average cost of a US data breach surpassed $10 million for the first time in 2025, as threat actors increasingly exploit unsecured AI systems and fragmented data architectures. Additionally, 35% of US organisations hit by breaches reported regulatory fines exceeding $250,000.

A carefully coordinated architecture ensures data is securely stored, processed, and governed consistently from end to end—especially critical in regulated sectors like financial services, government, and healthcare.

Strategic Partnerships Drive Success

Approximately 80% of organisations have outdated tech infrastructure requiring compute, storage, and network upgrades. Industry leaders are increasingly joining forces through strategic partnerships to unlock value impossible to achieve alone.

Collaboration has become the new standard for innovation and growth. Companies that embrace strategic partnerships whilst implementing hybrid-by-design infrastructure are best positioned to bridge the AI ROI gap.

Source: TechRadar

Share this article